基于聚类分析与几何的目标特征敏感性评估算法  

Target feature sensitivity evaluation method based on clustering analysis and geometry

作　　者：吴瑕[1] 周焰[2] 杨龙坡[2] 崔建[2] 

机构地区：[1]空军雷达学院雷达兵器运用工程军队重点实验室,武汉430019 [2]空军雷达学院预警监视情报系,武汉430019

出　　处：《控制与决策》2012年第6期914-918,共5页Control and Decision

基　　金：国家自然科学基金项目(40101019);空军雷达学院科研基金项目(LYYY200807)

摘　　要：为克服在较少先验知识的情况下,人为选择目标特征进行分类识别的盲目性,提出一种基于聚类分析与几何学的目标特征评估方法.该方法在类内类间距离的基础上引入几何学中有关圆的知识,定义了样本在特征空间中分布的"紧密度"、"分离度"函数和"松紧度"权重,以考察目标特征对目标样本之间的分类敏感性,并根据评估结果选取适当特征构成一个新的联合特征,以提高对目标的分类识别效率.最后通过仿真验证了该方法的有效性.To overcome the blindness of subjective selecting dimensionless indictors of target features as sensitive features with less experience,a target feature sensitivity evaluation method based on clustering analysis and geometry is proposed.The method defines ＂compact degree＂,＂dissociative degree＂ and ＂incompact-compact degree＂ of samples distributing in features space,and introduces the knowledge of rotundity from geometry,by which the classing ability of target feature to multi-class target samples is reviewed.Then the new combined features are constructed through the sensitivity evaluation results,and the classify recognition efficiency can be improved.Finally,simulation results show the effectiveness of the method.

关 键 词：聚类分析 几何学 类距离 特征评估 敏感性 目标识别 

分 类 号：TP181[自动化与计算机技术—控制理论与控制工程]